Google Apps Performance
Consultant Dashboard

🔶 About Me

I am a Doctor of Neuroscience with strong experience in data analysis, statistical modelling and research design. I focus on translating complex data into actionable insights for business and policy. I enjoy data wrangling, visualization, and project management.

Skills & tools:

• advanced R, advanced Python (pandas, NumPy, matplotlib, seaborn, scipy),
• developing my skills in Power BI and Power Apps,
• developing my skills in SQL (ETL, PostgreSQL, pgAdmin4, DBeaver),
• comfortable managing AI-augmented workflow, leveraging Cursor IDE and Claude,
• technical documentation in Jupyter Notebook (Markdown syntax), version control in Git.

  Contact: dominika.a.drazyk@gmail.com
  LinkedIn: in/dominika-drazyk-otw95

🔶 Project Navigation

Select the path that best matches your interest:

1. Executive & Business Insight
For reviewers focused on storytelling, strategy, and end-results.

• PDF Presentation: a step-by-step walkthrough of the project’s assumptions, technical execution highlights, and business insights;

• Images: a repository of example dashboard screenshots.

2. Technical Deep-Dive & Audit
For reviewers interested in the full analytical process and data interpretation.

• PowerBI Dashboard: an operational interactive dashboard;

• Data Preprocessing Code: production-ready, concise .py preprocessing code.

✳️ Dependency Management
This project uses Poetry to ensure a deterministic environment (locked versions) and 100% reproducibility. For basic users, a standard requirements.txt is also maintained.

• Option 1: Modern Workflow (Recommended) Use this if you have Poetry installed. This will automatically create a virtual environment and install the exact versions from poetry.lock.

# Bash
# Install dependencies and create virtual environment
poetry install
# Activate the environment
poetry shell

• Option 2: Standard Workflow (Pip) Use this for a traditional setup using the provided requirements.txt.

1. Initialize the Virtual Environment

# Linux / macOS
python3 -m venv .venv && source .venv/bin/activate
# Windows 
python -m venv .venv && .venv\Scripts\activate

2. Install Dependencies

# Bash
pip install --upgrade pip
pip install -r requirements.txt

🔶 Overview

Imagine you are a Market Research Consultant hired by a mobile app company. One of your tasks is to assist the Product Manager and evaluate their app ideas. Managers need fast, objective insight into how their concept fits the current mobile-app landscape.

This Consultant Dashboard provides exactly that. It helps the Consultant to:

• present the most competitive app categories;
• showcase the specific category targeted by the creators;
• compare their goals against market medians.

With these inputs, the app creators can ground their decisions in data rather than assumptions and quickly understand where their app stands relative to the market.

What this project delivers:

• A reproducible, well-documented merged datasets ready for dashboards or further analyses.

• Re-usable preprocessing code that can refresh results when source datasets are updated.

• PowerBI Consultant Dashboard – ready to use.

• 〽️ Business-relevant insights for mobile app creators, including market research and trend prognosis.

Objectives

• Prepare a clean, merged dataset from both Kaggle .csv files.
  Code: preprocessing_code.py

• Load the dataset into PowerBI and prepare a star schema model.
  File: Consultant_Dashboard.pbix

• Design drill-down hierarchy for Category-Genre levels, calculated columns and aggregation measures.
  File: Consultant_Dashboard.pbix

• Plan for multi-page thematic layouts with navigation and filtering panels.
  File: Consultant_Dashboard.pbix

• Create data visualizations, KPI cards and slicers.
  File: Consultant_Dashboard.pbix

• Make sure that dashboard design provides a functional market analysis flow useful to Product Managers.
  File: Consultant_Dashboard.pbix

Data & Source Metadata

External data source (Kaggle): Google Play Store Apps dataset Authored by L. Gupta (2019) and available under CC BY-SA 4.0.

Key variables

• App – the name of the mobile application.

• Category – the high-level classification assigned to the app (e.g., Tools, Games).

• Genre – more granular thematic type nested within each category.

• Star Rating – average user star rating.

• Count of Ratings – number of ratings submitted.

• Installs – total download count.

• Price in USD – app price in US dollars.

• Content Rating – age suitability classification (e.g., Everyone, Teen).

• Size in KB – application file size in kilobytes.

• Last Updated – date of the most recent app update.

• Count of Reviews – number of text reviews submitted.

• % of Positive Sentiment – the percentage of positive sentiment detected in a given text review.

• % of Neutral Sentiment – the percentage of neutral sentiment detected in a given text review.

• % of Negative Sentiment – the percentage of negative sentiment detected in a given text review.

• Sentiment Polarity – how polarized (Positive-Negative) was the sentiment detected in a given text review.

• Sentiment Subjectivity – how subjective was the narration of a given text review.

Tools & Methods

Data Preprocessing: Python {pandas, matplotlib, seaborn}

Data Analysis and Visualization: Dashboard – Microsoft PowerBI

Version control & sharing: Git & GitHub

🔶 Limitations & Challenges

• The Last Updated field does not extend beyond 2018, indicating that the source dataset is no longer current and may not fully reflect today’s app-store landscape.

• Several Genres contain only a small number of Apps, which limits the reliability of some category-level comparisons or trend-based interpretation.

❕ This dashboard was created as a portfolio project to demonstrate analytical and data-modelling skills, as well as the Power BI proficiency. While the dataset has clear limitations, it serves adequately for those purposes.

🔶 Presented skills

Data modelling

• Designing a star-schema model (fact table + dimension tables).
• Creating one-to-many and one-to-one relationships.
• Resolving ambiguity with bridging tables (App–Genre–Category).
• Creating calculated columns (for categorical ranges and ordering).
• Building aggregation measures (counts, medians, ratios).

Data visualization

• Designing charts with drill-down hierarchies (Category → Genre).
• Configuring scatterplots with logarithmic scaling for a clean data presentation.
• Creating user inputs and KPI cards with goal-vs-median logic.
• Creating multi-page thematic layouts with page navigation and filter panels.

Storytelling

• Translating a consultant-focused narrative into a functional dashboard.
• Highlighting high-impact apps visually.
• Organizing visuals into purposeful pages aligned with a consulting scenario.
• Preparing a PowerPoint to communicate methodology.

🔶 Portfolio Gallery

Project	Business Impact	Key Stack
euPOWERED Navigator	Policy-driven energy statistics & decarbonization tracking.	Python, Streamlit, Scraping
eCommerce Audit	Identifying financial risks and logistical bottlenecks.	PostgreSQL, ETL, Python
Consultant Dashboard	Market viability and competitive landscape engine.	Power BI, DAX, Star Schema
Efficiency and Diversity	Statistical research on human capital and spending.	Python, Scraping, Multivariate Stats

← Back to Profile